In a solid-state imaging device such as a CMOS image sensor, a problem of an SNR (i.e., S/N ratio) reduction arises due to a reduction of the amount of peripheral lights resulting from an optical lens. In a conventional imaging device, scanning for reset and scanning for read are varied in a frame in accordance with the quantity of flicker lights, and control is carried out so that a storage charge of each pixel becomes constant. In this way, a subject image having no reduction of image quality is obtained without depending on an electronic shutter speed under illumination having a flicker component such as a fluorescent lamp.